Title of Invention	A METHOD TO GENERATE EARTHQUAKE EARLY WARNING USING MEMS BASED SEISMIC SENSORS AND SYSTEM THEREOF
Abstract	The present invention relates to a novel Micro-Electro-Mechanical System based electronic earthquake early warning device that will generate an early earthquake warning signal before the arrival of destructive seismic waves, said device comprising: one or more monitoring systems composed of MEMS based accelerometer sensors; a processing unit that converts data to information; a warning issuing and communication system and real-time communication link that transmits data from the one or more of the said monitoring systems to the processing unit and from the processing unit to the warning issuing and communicating system.

Title of Invention

A METHOD TO GENERATE EARTHQUAKE EARLY WARNING USING MEMS BASED SEISMIC SENSORS AND SYSTEM THEREOF

Abstract

The present invention relates to a novel Micro-Electro-Mechanical System based electronic earthquake early warning device that will generate an early earthquake warning signal before the arrival of destructive seismic waves, said device comprising: one or more monitoring systems composed of MEMS based accelerometer sensors; a processing unit that converts data to information; a warning issuing and communication system and real-time communication link that transmits data from the one or more of the said monitoring systems to the processing unit and from the processing unit to the warning issuing and communicating system.

Full Text	Field of the Invention: The present invention relates to a novel Micro-Electro-Mechanical System based electronic earthquake early warning device that will generate an early earthquake warning signal before the arrival of destructive seismic waves. Definitions, Background and Prior Art Description: In the following paragraphs, the following abbreviations / Acronyms have been used by the Applicant, which are intended to convey the meaning as defined here below unless specifically mentioned: MEMS:- Micro-Electro-Mechanical Systems (MEMS) is the integration of mechanical elements, sensors, actuators, and electronics on a common silicon substrate through micro-fabrication technology. P-Wave:- Primary (P) Wave is the first wave radiated out when an earthquake occurs. The first arrival is the primary or P waves which usually have relatively low-amplitude and causes little damage. P-waves are longitudinal waves, S-Wave:- Secondary wave or Shear wave immediately follows P-wave. The S-wave usually has larger amplitude. S-waves are transverse in nature and exhibit shear vibration in a plane perpendicular to the direction of propagation. Thus, as compared to primary waves, secondary waves are more damaging. Surface Waves: Surface Waves originate on the surface of the earth (the epicenter) as compared to P and S waves which originate at the source to the earthquake underground (focus). Surface waves travel even slower than S waves (approximately 2 times slower than the S-waves) and are extremely dangerous because they posses both vertical and horizontal components. Seismic Waves: The term "Seismic waves" include both the Secondary waves (S-waves) as well as the surface waves. MSS: - MEMS based Seismic Sensor. When an earthquake occurs, energy radiates in all directions as three types of seismic waves called primary, secondary and surface waves. The energy of primary waves (P- waves) travels through the earth as a sequence of back and-forth vibrations parallel to the direction of propagation of the seismic waves. Secondary waves (S-waves), also called shear waves, are transverse in nature. The principle of separating P waves from S waves is as follows: • the P waves travel approximately 1.7 times faster than the S waves, • the greater the distance from the focus of an earthquake one is, the greater would be the time elapsed between the P and S waves. Based on this, the P waves can be separated from the S waves. Surface waves (Rayleigh waves and Love waves) are the slowest and most destructive among the three. Surface waves are approximately 2 times slower than the S-waves. Surface waves differ nom P and S waves because they begin on the surface of the earth (the epicenter) as compared to P and S waves which originate at the source to the earthquake underground (focus). Surface waves travel even slower than S waves and are extremely dangerous because they posses both vertical and horizontal components. Thus by detecting P-waves and/or S-waves we can generate an Early warning at least moments before the actual shaking of a structure. The Early warning signal can be used to provide opportunities to automatically trigger measures such as shutdown computers; reroute electrical power; shutdown disk drives; shutdown high precision facilities; shutdown airport operations; shutdown manufacturing facilities; stop trains; shutdown high energy facilities; shutdown gas distribution; alert hospital operating rooms; open fire station doors; start emergency generators; stop elevators in a safe position; shutoff oil pipelines; issue audio alarms; shutdown refineries; shutdown nuclear power plans; shutoff water pipelines; maintain safe-state in nuclear facilities. Statement of the present invention The present invention is related to a method to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said method comprising steps of; positioning one MSS each horizontally and vertically onto a L-shaped platform at a predetermined distance from target station, detecting and processing P & S wave signals by said MSS, and sending early warning signal to the target station; and A system to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said system comprises; MSS positioned both horizontally and vertically onto a L-shaped platform to process P & S wave signals at a predetermined distance from target station, and means to send early warning signal to the target station. Objects of the Present Invention: The main object of the present inventions is to develop a method and a system to generate earthquake early warning signal using MEMS based seismic sensors (MSS). Yet another object of the present invention is to positioning one MSS each horizontally and vertically onto an L-shaped platform at a predetermined distance from target station. Still another object of the present invention is detecting and processing P & S wave signals by said MSS. Still another object of the present invention is sending early warning signal to the target station. Brief description of the Accompanying Drawings: In the drawings accompanying the specification, Figure 1 illustrates the block diagram of the MSS earthquake early warning device of the present invention. Figure 2 illustrates the block diagram of Example 1 wherein the MSS is placed away from the target station. Figure 3 illustrates the block diagram of Example 2 wherein the MSS is placed at the target station. Figure 4 shows typical duty cycle output of ADXL202 Figure 5 illustrates positioning of MSS on an L-shaped platform. Figure 6 shows the flowchart of DSP process. Figure 7 shows the flowchart for user interface with DSP. Figure 8 shows a schematic of the MSS hardware. Figure 9 shows a typical user interface. Detailed Description of the Present Invention: The present invention is related to a method to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said method comprising steps of; positioning one MSS each horizontally and vertically onto a L-shaped platform at a predetermined distance from target station, detecting and processing P & S wave signals by said MSS, and sending early warning signal to the target station. In another embodiment of the present invention wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock. In yet another embodiment of the present invention detecting P & S waves using accelerometer. In still another embodiment of the present invention wherein the accelerometer generates Pulse Width Modulated (PWM) output. In another main embodiment of the present invention a system to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said system comprises; a) MSS positioned both horizontally and vertically onto a L-shaped platform to process P & S wave signals at a predetermined distance from target station, and b) means to send early warning signal to the target station. In yet another embodiment of the present invention the L-shaped platform is fixed onto a concrete vault built over hard soil or rock. In still another embodiment of the present invention the sensor is an accelerometer. In still another embodiment of the present invention the MSS comprises a Digital Signal Processor (DSP). In still another embodiment of the present invention the system comprises to store the processed data. The present invention relates to the product realization, using MEMS based accelerometer, of a novel concept for generating earthquake early warning before the arrival of destructive seismic waves. MSS is a MEMS based electronic earthquake early warning device that will generate a warning signal before the arrival of destructive seismic waves. MEMS based seismic sensor is sensitive enough to detect small ground motions associated with strong earthquakes. It generates a warning signal before the arrival of more damaging motions produced by strong earthquakes. Early warning will help users to take precautionary action in the event of an upcoming seismic shaking. The warning time will depend, of course, on the earthquake type, the epicenter and focal distances, and the geological conditions of the site where the device is to be installed. MSS is a low cost, small size, fast responding embedded system which can detect the arrival of Earthquakes and generate alarm which can be used for shutting down vulnerable facilities before the arrival of the destructive surface waves. Design around MEMS based motion sensors and High speed processors that can more quickly distinguish the size of a quake and rapidly make a decision whether or not to shut off critical services. An Electrical signal is generated when the precursor of destructive strong motion earthquake is detected above pre set threshold. As can be noticed from Figure 1, the Micro-Electro-Mechanical Systems based Seismic Sensor for earthquake early warning consists of the following four modules: (1) one or more monitoring systems composed of MEMS based accelerometer sensors (2) a processing unit that converts data to information (3) a warning issuing and communication system and (4) real-time communication link that transmits data from the one or more of the said monitoring systems to the processing unit and from the processing unit to the warning issuing and communicating system. The processing unit which processes the data received into information comprises: • a high speed processor based signal processing card (MSS Signal computer card) and • a vertical accelerometer card (MSS Vertical card). MSS Signal computer card is specifically made to acquire and process the ground vibration signals using sensitive MEMS based accelerometers and high speed processor. The software for MSS consists of two modules, namely: a) Signal processing embedded system software running in processor b) User interface software running in PC Signal processing software running in processor is embedded system software, which acquires and processes the accelerometer data corresponding to ground vibrations and evokes the early warning alarm. User interface software running in PC provides the user interface facilities for communicating with MSS. DSP software The main objective of this program is to acquire 2 seconds real-time PWM data of vertical and horizontal channels of ADXL202 accelerometers and process the ground acceleration data to detect P-wave and S wave. The Richter magnitude (Ml) and epicenter distance of possible earthquake is calculated using early warning algorithm. If the estimated Richter scale magnitude crosses the pre set threshold (Ml) then a warning is generated. Five sub modules namely, Initialization routines, Set up parameters, Data Acquisition, Data Processing and Warning generation 125 are included which are designed around serial and timer interrupts of DSP. Data Processing is done in every 2 second. Serial interrupt is active when the user wants to send initialization parameters from local computer to MSS. Timer interrupt is active whenever the accelerometer PWM data is received at the timer ports of the processor. Data acquisition is initiated based on timer interrupt. Data processing is done in the main body of the program. Initialization routines take care of the initial MSS health checkups, Set up parameter sub module looks after the receipt of user selectable initialization parameters and passing of parameters to processor. Warning generation module evokes the early warning if Data processing module detects an event above the threshold. The flow chart of DSP software is given in figure 6. Sub module 1-Initialization routines Immediately after power up initialization module takes the control of code execution and checks the health status of MSS by inspecting the DSP interfaces with internal memory, external memory and external interfaces. The result of the initialization test is displayed by LED 2 to 3 of MSS using the flag I/O pins (FLAG 0-1). At the end of this module serial interrupt is enabled. The status of ConfigEdit flag is checked to either run with default values of set up parameters or accepting values from PC. Sub module 2-Set up parameters User can send the initialization parameters like values for Threshold strong motion, lower value of Tp, Upper value of Tp, S-wave detection factor, Range for processing, Valid count for frame validation and Threshold value of Ml to MSS from local computer user interface. After the execution of initialization routines sub module, processor enters to a for ever loop waiting for interrupts to come. Once the set up parameters are sent from local computer to the serial port of DSP in MSS, Interrupt Service Subroutine (ISR) for serial interrupt is evoked. In the ISR for serial data reception, the received ASCII characters are stored to a specific memory location after converting to float values and the same characters were sent back to local computer as an acknowledgement. If the character received at MSS is a valid one (valid means any of ASCII character 0 to 9, A to Z or decimal point) it is read from the receive buffer and stored in a receive array. This process is continued till all parameters are received from local computer. If the received character is 'Y' previous digits received are taken as the parameter for Threshold values for percentage strong motion. Similarly arrival of character 'X' indicates lower value of Tp, 'W indicates upper value of Tp,'V indicates S-wave detection factor, 'U' indicates range of processing after P-wave detection, 'T' indicates valid count for frame validation, 'K' indicates threshold for Ml. and arrival of character 'Z' for end of reception of parameters. The received parameters are then transferred to corresponding variables. Finally serial interrupt is disabled, timer interrupt is enabled and processor goes to forever wait loop waiting for timer interrupts Calibration If the character received from PC is 'C then MSS is delegated to calibration mode by setting calibration flag and ten second sensors data are acquired and from these values Average static acceleration values in X(g) and Y(g) is calculated and these values are send to PC . The user can either accept these values so that DC offset removal is done by using these values or reject to run MSS with default DC offset removal values. Sub module 3 - Data Acquisition Data acquisition module consists of the ISR of timer interrupt. Horizontal and Vertical accelerometer PWM data are interfaced to Timer 1 and Timer 0 of DSP respectively. Whenever a PWM signal is received at the timer ports of DSP after enabling timer interrupt, the program will jump to the ISR of timer interrupts. In the timer ISR, the ON period (Tl) and total period (T2) of PWM signal taken from TCOUNT and TPERIOD registers respectively of timer. The duty ratio(Tl/T2) is calculated and stored in circular buffer, A TICKS (real-time system clock) counter is incremented on receipt of each set of PWM data. When the value of TICKS counter reaches twice the sampling frequency (equals to 2 second data), the program sequence goes to Data processing module. Sub module 4 - Data processing Data acquired by the Data Acquisition module are being processed in this module. The ground acceleration for both channels are calculated as follows: - Acceleration in g = (Duty ratio - 0.5)/12.5 %; Since 0 g corresponds to 50 % Duty Cycle of PWM signal of accelerometer The acceleration values are stored to acceleration buffer. Two second acceleration data samples of both x and y axis data is stored in process buffers and is read for further processing. The TICKS counter is reset to zero value. The following processes are carried out on the samples:- a) DC offset removal This process is necessary because in acceleration readings earth's g value and Og offset errors will act as a DC offset. DC shift is removed by subtracting either the average static g values calculated in calibration mode or the default values set by the program. This is done by using the following formula. New value of ith sample value= ith g value -De value This is applicable for both x and y-axis. b) Apply an FIR Low pass Filter with pass band end frequency 2Hz and stop band start frequency 5Hz to filter out extraneous noise. c) Energy (Mean square value of one frame of data) is calculated before and after filtering. Energy passed through filter is calculated from this energy for both X and Y-axis. d) Frame validation is done for X axis data for rejecting noise pick ups below noise floor and above the upper limit of accelerometer output saturation Validation is done for X-axis data alone since arrival of P-Waves to be detected from vertical accelerometer output. e) Maximum g value and number of g that crosses the set value (Percentage strong motion) is found from both axis data. f) Tp, Predominant Period is calculated from validated X-axis data samples, if energy passed is greater than preset energy threshold (default value 40%). Algorithm for finding Tp Where TjP is the predominant period at time i, aj is the recorded ground acceleration , Ai is the smoothed ground acceleration squared, Jj is the jerk(time derivative of acceleration) squared and a is the smoothening constant .The value of a is 0.99. g) The maximum value (Tp-max) of predominant period Tp, from 2-second samples are found out. When Tp-max is between lower value of Tp and upper value of Tp, then P- wave is detected following parameters are calculated. mh, Body Magnitude = 7.0 * Logio(Tp-Max) + 5.9 Ml, estimated Richter magnitude = (1.5873*mh)-3.968253968 P-wave Confidence Factor is incremented whenever a P wave is detected. Process to Detect S-wave h) Once a P- wave is detected and the process will search for an S-wave from both X and Y-axis data. An S-wave is detected when current frame energy is >= twice that of detected P wave. When S- wave is detected, S-Flag is set and the time interval between P- wave and S-wave arrival is calculated. This time interval multiplied by 8KM will give the epicenter distance since there is typically a one second (approximately) separation between the P and S waves for every 8 KM traveled. This process is repeated for next frames of data. 1.4.Sub module S - Sending Results Calculated parameters are sent from MSS to user interface running in local computer through the serial port of DSP. It includes the following: - 1) Result of Calibration 2) Maximum 'g' value of each frame. 3) Percentage Strong Motion 4) X (g) and Y (g) values 5) Maximum Tp. 6) Primary Confidence Factor 7) Secondary Confidence Factor 8) Ml, predicted Richter Magnitude 9) P-S time User interface software MSS User interface is a software developed using CVI(C for Virtual Instrumentation) to run in a local computer in order to communicate with MSS. The main functions carried out by this program are sending initialization parameters to MSS, displaying results of processing and to store the data and results of processing in files. Parameters like threshold strong motion, Tp-Lower (Lower value of Predominant Period), Tp-Upper (Upper value of Predominant Period), S-Wave Detection Factor, Range, Valid Count and Threshold Ml are send to the MSS via the serial port of PC/Laptop by pressing the SEND button in the user interface. Identification tag is attached with every parameter to distinguish between them .At the receiving side values of parameters are stored in corresponding memory location for processing. MSS transmits the received values back to Laptop in order to make sure that the send parameters were reached in the destination properly. The CVI program running on the PC stores and displays the results sent by MSS. When the CALIBRATE button is pressed the system enters in calibration mode by acquiring ten second acceleration data in both x and y-axis for calculating average static g values X (g) and Y (g) for DC offset removal. The calculated values are displayed in the acknowledgement window, user can either accept these values by pressing ACCEPT button or can run with default values set by MSS program by pressing RUN button. MSS can be tested with simulated signals from programmable arbitrary wave form generator (Eg:- Agilent 33220A Model).In that mode DC offset removal is not required. LABTEST button is clicked for running with simulated signal. When RUN button is clicked serial interrupt is disabled, timer interrupt is enabled and processor goes to a 'for-ever-wait' loop waiting for interrupts and continuously acquires data and processes it and the results are transmitted through the serial interface. A real time data plotting facility is provided for displaying x and y channel acceleration data. Data and results are also stored in separate files, thus the user can analyze the results in offline. In the user interface 'P wave detected', 'S wave detected' message and calculated values of maximum 'g' in x and y channel data, percentage strong motion value, maximum Predominant Period (Tp), confidence factor for P-wave and S-wave, Body Magnitude of Earthquake (mh), and equivalent Richter scale magnitude(Ml) are displayed. Once P- wave is detected, MSS looks for consecutive detection of P-wave to set the confidence level and the arrival of S-wave. When S-wave is detected distance from Epicenter and confidence level is sent by MSS is displayed on Local computer. The flow chart of the program is as shown in figure 7. User interface The user interface part of this program includes the following components. Numeric: - This control is used for displaying the values of maximum 'g' value, percentage strong motion, Ml, Ms, Tp-Max, Confidence factor for P-wave & S-wave and distance from epicenter. Text Message: - Text Message is used to display strings of text. When a P-wave is detected it shows 'P-wave detected' message and when S wave is detected it shows 'S-wave detected' message. Command Button: - This control is used to trigger an action. SEND button associated with this is used for sending the parameters to MSS. CALIBRATE button is used for displaying results after calibration. ACCEPT button is used for accepting the results of calibration RUN button is associated with running in continuous mode. LABTEST button is used to run without DC offset removal EXIT button is used for closing COM port, files and to quit the user interface Ring Control: - Ring control is used to select from a group of items. Each item is stored in a ring control in the form of label/value pairs where the label is the text string displayed by the control when the item is selected and value is numeric or string value corresponding to that item. When this control is selected the user can set the parameter values for Threshold strong motion, Tp-Lower (Lower value of Predominant Period), Tp-Upper (Upper value of Predominant Period), S-Wave Detection Factor, Range, Valid Count and Threshold Ml. Textbox Control: - This control is used for displaying acknowledgement and message to user. Strip Chart: - Strip Chart control display graphical data in real time, consisting of one or more traces that is updated simultaneously. This control is used for displaying x and y axis acceleration data in real time. A typical user interface is as shown in figure 9. MSS is a digital signal processing embedded system designed to acquire and process the ground vibration signals using sensitive MEMS based accelerometers and high speed DSP. The hardware consists of a DSP based signal processing card (MSS Signal computer card) and a vertical accelerometer card (MSS Vertical card). DSP will continuously acquire the signals from the accelerometers through its timers and will process it based on early warning algorithms for earthquake. A suitable electrical signal is generated when the processed magnitude crosses pre set threshold. The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. The processor used is ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS and 544Kbits internal SRAM. Hardware-MSS Signal Computer card & MSS Vertical card MSS Signal computer card specifically designed to acquire and process the ground vibration signals make use of sensitive MEMS based accelerometers and high speed DSP. MSS Vertical card is based on sensitive MEMS based accelerometer to measure the vertical ground vibrations. The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. This card is fixed on the MSS signal computer card vertically so that the vertical vibrations of earth could be captured. The picture of MSS hardware is shown in figure 8. The block schematic of the MSS hardware is shown figure 1. According to the block schematic of the MSS hardware is shown figure 1, the accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. It is a low power, low cost device having a resolution of 2mg at 60Hz.Two dual axis accelerometers are used to sense the horizontal and vertical vibrations. Horizontal accelerometer is fixed on MSS Signal computer card and vertical accelerometer is placed on MSS vertical card which is fixed vertically to MSS Signal computer card. The processor selected is Analog Device's ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS. An external Synchronous DRAM memory with 256Mbits is provided for storing program data like look up tables etc. An EPROM with 256KB size is provided to keep the program code. JTAG interface is provided to communicate with DSP during the software development cycle. Reset circuitry ensures the proper processor reset to a known state and ensures the execution of code from the hardware vectored address of program memory. The external communication interface of MSS is realized through the dedicated serial port of DSP. Accelerometer Selection The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. It contains a polysilicon surface-micro machined sensor and signal conditioning circuitry to implement open loop acceleration measurement architecture. It is a low power, low cost device having a resolution of 2mg at 60Hz. The major features of ADXL202JE are:- • MEMS based 2-axis acceleration sensor on a single chip • 5 mm x 5 mm x 2 mm ultra small chip scale package • Direct interface to DSP timer ports via Duty Cycle output • Band Width adjustment with a single capacitorO ADXL202JE can measure both dynamic acceleration (E.g. vibration) and static acceleration (eg. gravity). Two dual axis accelerometers are used to sense the acceleration in three mutually perpendicular axes. The duty cycle (ratio of pulse width to period) of PWM output of the sensor (figure 4) is proportional to acceleration. DSP timer ports can directly decode the duty cycle outputs. Acceleration in g = ((T1/T2) - 0.5)/12.5 % 0 g = 50 % Duty Ratio (T1/T2) T2(sec) = RSET(^ )/125MQ Where RSET is an external resistor to the ADXL202 Accelerometer. Typical Duty Cycle output of ADXL 202 is as shown in figure 4. The MSS senses the arrival of primary (P) waves, the first waves to reach the detector, and so provides adequate time to react before the arrival of more destructive shear (S) and surface (Love and Rayleigh) waves. The MSS will filter out extraneous noise components. Accelerometer interface to DSP The Pulse Width Modulated output of ADXL202 is directly interfaced to the timer ports of Analog Devices' DSP ADSP 21065L.One sensor is placed on the MSS Signal computer card which contains the DSP while the other sensor is fixed on the smaller MSS vertical card which is soldered vertically to the MSS Signal computer card. From the two accelerometers four channels are available-namely, Horizontal X axis, Horizontal Y axis, Vertical X axis and Vertical Y axis. Processor- ADSP 21065L The processor selected is ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS from Analog Devices. It has four independent buses for Dual data, instructions, and I/O fetch on a single cycle. The key features of ADSP21065Lare:- • 32-Bit Fixed-Point Arithmetic; 32-Bit and 40-Bit Floating Point Arithmetic • User-Configurable 544 Kbits On-Chip SRAM Memory • 66 MIPS, 198 MFLOPS Peak, 132 MFLOPS sustained performance • Two External Port, DMA Channels and Eight Serial port • SDRAM Controller for Glueless Interface to Low Cost external memory (@ 66 MHz) • 12 Programmable I/O Pins and Two Timers with Event Capture Options • Serial Ports with independent Transmit and Receive Functions • 3.3 Volt Operation Internal SRAM memory size is 544Kbits and an in built external memory interface is provided. Signal from each axis of accelerometer is acquired and processed by the DSP through its timer ports-Timer 0 and Timer 1. DSP will continuously acquire the signals from the accelerometers through its timers and will process it based on early warning algorithms for earthquakes. DSP-External Memory interface ADSP 21065L has in built SDRAM controller by which glue less interface of SDRAM (Synchronous Dynamic RAM) is achieved. An external Synchronous DRAM memory with capacity of 128Mbits from Micron Technology (Part No.MT48LC8M16A2TG-75) with 133MHz maximum refresh rate is used for storing program data like look up tables etc. and processed outputs of the sensor which crosses the threshold. Two SDRAMs are used so that a total memory capacity of 256Mbits is ensured. An EPROM with 256KB size is provided to store the program code. This EPROM can be made as the boot device by hardware selection provided. Memory size selection jumpers are provided so that the EPROM size can be increased up to 1MB with same pin out if required. Setting the BSEL input of DSP high and the BMS input low selects EPROM booting through the external port. The data pins of the byte-wide boot EPROM is connected to DATA7-0 of DSP. The lowest address pins of the processor are connected to the EPROM's address lines. The EPROM's chip select is tied to BMS and its output enable to RD signal of DSP DSP Clock circuitry Stabilized clock input to DSP and external devices are ensured using a programmable clock generator having integrated phase-locked loop and frequency select option. CY2292F from Cypress semiconductors is a 3.3V EPROM programmable clock generator with input crystal frequency varies from 10MHz-25MHz and programmable output clock frequency from 76.923 kHz to 66.6MHz. In this design an external 12MHz crystal is used as the input source. CY2292F is programmed to output 33MHz (DSP_CLK1) and 66MHz (DSP_CLK2) clock outputs DSP Reset Circuitry Reset circuitry for DSP ensures the proper processor reset to a known state and ensures the execution of code from the hardware vectored address of program memory. 3.3V Reset IC from Dallas Semiconductors, DS1233A, is used as the Reset IC which provides a 350ms Reset Pulse during power on reset as well as externally forced reset conditions. DSP JTAG Interface JTAG interface is provided to communicate with DSP during the software development cycle using Analog Devices' USB based In Circuit Emulator. Compiled output of codes can be downloaded to the internal memory of DSP through JTAG interface. JTAG based In Circuit Emulator helps to debug the software codes running in DSP. External Communication Interface The external communication interface of MSS is realized through the dedicated serial port of DSP. RS232 link is provided for communicating with local computer and an optional RF/optical fiber link can be added for communication with other MSS and central computer. Using the synchronous serial ports (SPORTS) on the SHARC DSP, it is possible to implement a full-duplex, asynchronous serial port to communicate with UARTs, EIA-232 (RS-232) devices with minimal software overhead (less than 1 MIP for full-duplex, 115,200 bps operation) The prototype of MSS was extensively tested with simulated signals at CDAC lab and in simulated vibration environment at ERTL lab, Trivandrum for validating the Early Warning algorithm. MSS was also subjected to noisy environment tests to prove that false alarms due to local noise sources are eliminated. MSS was also tested using recorded strong motion seismograms available in public domain. We visited and contacted through mail to the following sites www.pnsn.org, www.anss.om, www.usgs.gov etc. We downloaded recorded digital data for strong motion earthquake from above mentioned sites and tested MSS. We also tested MSS in following environments:- • Using the vibration test facility at ERTL, Thiruvananthapuram with frequencies 1Hz, 2Hz , 3Hz , 5Hz and 10Hz with acceleration as low as 20millig. The noise pickups at lower range of vibration table limited the test for low value of acceleration and lower frequency of vibration • Using simulated PWM signal corresponding to P-S wave using Agilent's arbitrary wave form generator • At quarry during rock blasts • Near railway track when train passes • Near National High-way during heavy vehicle traffic By these tests the early warning algorithm is validated. Tests conducted near noise sources proved that MSS is not generating false alarms. The MSS has to be installed properly for field applications in such a way that the natural frequency of the mechanical set up is well outside the band of P wave and S wave and the ground vibrations are coupled to MSS with minimum attenuation. Also the possible entry of ambient noises such as vehicle traffic, wind, rain etc. to MSS setup should be minimized. For permanent fixture of MSS following setup can be used. The structure may consist of an L-shaped platform fixed to a concrete vault built on hard soil or rock as shown in figure 5. MSS is fixed firmly on the mounting platform. To prevent noise from local sources like vehicle traffic guard the vault using loose sand .MSS should also be protected from wind, dust, sun and rain with proper roof and cover. The current prototype of MSS is designed for two-axis measurement of seismic signals. Earthquake early warning system requires three-axis measurement of seismic signals. Therefore two MSS modules have to be used. The fixing arrangement of two MSS modules is as shown in figure 5. One MSS module (consists of MSS Signal Computer Card and MSS Vertical card) has to be fixed horizontally. The other MSS module (consists of MSS Signal Computer Card and MSS Vertical card) has to be fixed vertically. MSS detects the pre-cursor of a strong earthquake and generates an early warning signal as follows. a) Generate an Electrical Early warning signal based on the characteristics of P wave only. b) Generates a warning based on the analysis of P-S complex waves. MSS can also be used as a low cost low sensitivity Seismograph. MSS Performance Characteristics: • Generate warning signal for Magnitude> 5.5 (can be set) • Fast processing: less than 50 milliseconds (can be reduced) • Operation: 3.3V, 0°C to 50°C (can be improved) • Power consumption: 600milli watts (can be reduced) • Time taken to generate Early warning: less than 2.5 sec from P-wave arrival (can be reduced) The following examples are provided purely by way of exemplification and illustration and should not be construed to limit the scope of the invention in any manner. The invention is further elaborated with the help of following examples. However these examples should not be construed to limit the scope of the invention. Examples Determination of Expected Early Warning Time In the following two examples, the Applicants have made estimated time period which would be available for people at the target site to prepare for the earth quake after receipt of the early warning from the Micro-Electro-Mechanical Systems based electronic earthquake early warning device of the present invention. Example 1: When MSS is placed away from Target station When an earthquake occurs, energy radiates in all directions as three types of seismic waves called primary, secondary and surface waves. The energy of primary waves (P-waves) travels through the earth as a sequence of back and-forth vibrations parallel to the direction of propagation of the seismic waves. Secondary waves (S-waves), also called shear waves, are transverse in nature. . The principle of P and S wave separation is as follows. Since the P waves travel approximately 1.7 times faster than the S waves, the greater the distance from the focus of an earthquake, the greater would be the time elapsed between the P and S waves. There is typically a one-second separation between the P and S waves for every 8 km traveled. Surface waves (Rayleigh waves and Love waves) are the slowest and most destructive among the three. Assume MSS is placed fD' Kms away from the target station to be protected. Let 'd' be the distance of MSS from the epicenter as shown in figure 2. Time taken to generate warning signal at MSS from the arrival of P-wave = 2.2 sec Time taken to transmit warning to target station = 0.3 sec. Time taken to receive early warning signal at target station from the arrival of P-wave at MSS = P = 2.2 + 0.3 = 2.5 seconds. The velocity of the P-Wave depends on the type of soil .It varies from 0.3km/s in aerated soil to 1 lkm/s at the centre of the earth. In the upper mantle the velocity of the P Wave is 8km/s. Following derivation assumes that average velocity of P-Wave is 8km/s. Early Warning Time [TE] = (0.21D + 0.085d - P) seconds Hence the Early Warning Time [TE] = (0.21D + 0.085d - P) seconds. If MSS is installed 20KM from station to be protected then D=20KM, then for various 'd' Early warning time can be calculated as follows and the same is tabulated in Table 1: Example 2: When the MSS is located at the Target Site Assume MSS is placed at the target station to be protected. Let 'd' be the distance of MSS from the epicenter as shown in figure 3. Time taken to generate warning signal at MSS from the arrival of P-wave = P= 2.2 sec. Hence the Early Warning Time [T E] = (0.0850d-P) seconds. For various'd' Early warning time can be calculated as follows and the same is tabulated in Table 2: TABLE 2: EARLY WARNING TIME AVAILABLE Industrial Application: • In Control Systems for industries handling flow of hazardous and flammable material through pipes. • Seismic Fence for Atomic and Thermal Power stations. • As low sensitivity low cost Seismograph. • It provides opportunities to automatically trigger measures such as shutdown computers; reroute electrical power; shutdown disk drives; shutdown high precision facilities; shutdown airport operations; shutdown manufacturing facilities; stop trains; shutdown high energy facilities; shutdown gas distribution; alert hospital operating rooms; open fire station doors; start emergency generators; stop elevators in a safe position; shutoff oil pipelines; issue audio alarms; shutdown refineries; shutdown nuclear power plans; shutoff water pipelines; maintain safe-state in nuclear facilities. We Claim: 1. A method to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said method comprising steps of; a. positioning one MSS each horizontally and vertically onto a L-shaped platform at a predetermined distance from target station, b. detecting and processing P & S wave signals by said MSS, and c. sending early warning signal to the target station. 2. The method as claimed in claim 1, wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock. 3. The method as claimed in claim 1, wherein detecting P & S waves using accelerometer. 4. The method as claimed in claim 3, wherein the accelerometer generates Pulse Width Modulated (PWM) output. 5. A system to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said system comprises; a. MSS positioned both horizontally and vertically onto a L-shaped platform to process P & S wave signals at a predetermined distance from target station, and b. means to send early warning signal to the target station. 6. The system as claimed in claim 5, wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock. 7. The system as claimed in claim 5, wherein the sensor is an accelerometer. 8. The system as claimed in claim 5, wherein the MSS comprises a Digital Signal Processor (DSP). 9. The system as claimed in claim 5, wherein the system comprises to store the processed data.

Full Text

Field of the Invention:
The present invention relates to a novel Micro-Electro-Mechanical System based electronic earthquake early warning device that will generate an early earthquake warning signal before the arrival of destructive seismic waves.
Definitions, Background and Prior Art Description:
In the following paragraphs, the following abbreviations / Acronyms have been used by the Applicant, which are intended to convey the meaning as defined here below unless specifically mentioned:
MEMS:- Micro-Electro-Mechanical Systems (MEMS) is the integration of mechanical elements, sensors, actuators, and electronics on a common silicon substrate through micro-fabrication technology.
P-Wave:- Primary (P) Wave is the first wave radiated out when an earthquake occurs. The first arrival is the primary or P waves which usually have relatively low-amplitude and causes little damage. P-waves are longitudinal waves,
S-Wave:- Secondary wave or Shear wave immediately follows P-wave. The S-wave usually has larger amplitude. S-waves are transverse in nature and exhibit shear vibration in a plane perpendicular to the direction of propagation. Thus, as compared to primary waves, secondary waves are more damaging.
Surface Waves: Surface Waves originate on the surface of the earth (the epicenter) as compared to P and S waves which originate at the source to the earthquake underground (focus). Surface waves travel even slower than S waves (approximately 2 times slower than the S-waves) and are extremely dangerous because they posses both vertical and horizontal components.
Seismic Waves: The term "Seismic waves" include both the Secondary waves (S-waves) as well as the surface waves.
MSS: - MEMS based Seismic Sensor.
When an earthquake occurs, energy radiates in all directions as three types of seismic
waves called primary, secondary and surface waves. The energy of primary waves (P-

waves) travels through the earth as a sequence of back and-forth vibrations parallel to the direction of propagation of the seismic waves. Secondary waves (S-waves), also called shear waves, are transverse in nature.
The principle of separating P waves from S waves is as follows:
• the P waves travel approximately 1.7 times faster than the S waves,
• the greater the distance from the focus of an earthquake one is, the greater would be the time elapsed between the P and S waves.
Based on this, the P waves can be separated from the S waves.
Surface waves (Rayleigh waves and Love waves) are the slowest and most destructive among the three. Surface waves are approximately 2 times slower than the S-waves. Surface waves differ nom P and S waves because they begin on the surface of the earth (the epicenter) as compared to P and S waves which originate at the source to the earthquake underground (focus). Surface waves travel even slower than S waves and are extremely dangerous because they posses both vertical and horizontal components.
Thus by detecting P-waves and/or S-waves we can generate an Early warning at least moments before the actual shaking of a structure. The Early warning signal can be used to provide opportunities to automatically trigger measures such as shutdown computers; reroute electrical power; shutdown disk drives; shutdown high precision facilities; shutdown airport operations; shutdown manufacturing facilities; stop trains; shutdown high energy facilities; shutdown gas distribution; alert hospital operating rooms; open fire station doors; start emergency generators; stop elevators in a safe position; shutoff oil pipelines; issue audio alarms; shutdown refineries; shutdown nuclear power plans; shutoff water pipelines; maintain safe-state in nuclear facilities.
Statement of the present invention
The present invention is related to a method to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said method comprising steps of; positioning one MSS each horizontally and vertically onto a L-shaped platform at a predetermined distance from target station, detecting and processing P & S wave signals by said MSS, and sending early warning signal to the target station; and A system to generate earthquake early warning signal using MEMS based seismic

sensors (MSS), wherein said system comprises; MSS positioned both horizontally and vertically onto a L-shaped platform to process P & S wave signals at a predetermined distance from target station, and means to send early warning signal to the target station.
Objects of the Present Invention:
The main object of the present inventions is to develop a method and a system to generate earthquake early warning signal using MEMS based seismic sensors (MSS).
Yet another object of the present invention is to positioning one MSS each horizontally and vertically onto an L-shaped platform at a predetermined distance from target station.
Still another object of the present invention is detecting and processing P & S wave signals by said MSS.
Still another object of the present invention is sending early warning signal to the target station.
Brief description of the Accompanying Drawings:
In the drawings accompanying the specification,
Figure 1 illustrates the block diagram of the MSS earthquake early warning device of
the present invention.
Figure 2 illustrates the block diagram of Example 1 wherein the MSS is placed away
from the target station.
Figure 3 illustrates the block diagram of Example 2 wherein the MSS is placed at the
target station.
Figure 4 shows typical duty cycle output of ADXL202
Figure 5 illustrates positioning of MSS on an L-shaped platform.
Figure 6 shows the flowchart of DSP process.
Figure 7 shows the flowchart for user interface with DSP.
Figure 8 shows a schematic of the MSS hardware.
Figure 9 shows a typical user interface.

Detailed Description of the Present Invention:
The present invention is related to a method to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said method comprising steps of; positioning one MSS each horizontally and vertically onto a L-shaped platform at a predetermined distance from target station, detecting and processing P & S wave signals by said MSS, and sending early warning signal to the target station.
In another embodiment of the present invention wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock.
In yet another embodiment of the present invention detecting P & S waves using accelerometer.
In still another embodiment of the present invention wherein the accelerometer generates Pulse Width Modulated (PWM) output.
In another main embodiment of the present invention a system to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said system comprises;
a) MSS positioned both horizontally and vertically onto a L-shaped platform to process P & S wave signals at a predetermined distance from target station, and
b) means to send early warning signal to the target station.
In yet another embodiment of the present invention the L-shaped platform is fixed onto a concrete vault built over hard soil or rock.
In still another embodiment of the present invention the sensor is an accelerometer.
In still another embodiment of the present invention the MSS comprises a Digital Signal Processor (DSP).
In still another embodiment of the present invention the system comprises to store the processed data.

The present invention relates to the product realization, using MEMS based accelerometer, of a novel concept for generating earthquake early warning before the arrival of destructive seismic waves.
MSS is a MEMS based electronic earthquake early warning device that will generate a warning signal before the arrival of destructive seismic waves.
MEMS based seismic sensor is sensitive enough to detect small ground motions associated with strong earthquakes. It generates a warning signal before the arrival of more damaging motions produced by strong earthquakes. Early warning will help users to take precautionary action in the event of an upcoming seismic shaking. The warning time will depend, of course, on the earthquake type, the epicenter and focal distances, and the geological conditions of the site where the device is to be installed.
MSS is a low cost, small size, fast responding embedded system which can detect the arrival of Earthquakes and generate alarm which can be used for shutting down vulnerable facilities before the arrival of the destructive surface waves. Design around MEMS based motion sensors and High speed processors that can more quickly distinguish the size of a quake and rapidly make a decision whether or not to shut off critical services. An Electrical signal is generated when the precursor of destructive strong motion earthquake is detected above pre set threshold.
As can be noticed from Figure 1, the Micro-Electro-Mechanical Systems based Seismic Sensor for earthquake early warning consists of the following four modules:
(1) one or more monitoring systems composed of MEMS based accelerometer sensors
(2) a processing unit that converts data to information
(3) a warning issuing and communication system and
(4) real-time communication link that transmits data from the one or more of the said monitoring systems to the processing unit and from the processing unit to the warning issuing and communicating system.
The processing unit which processes the data received into information comprises: • a high speed processor based signal processing card (MSS Signal computer card) and

• a vertical accelerometer card (MSS Vertical card).
MSS Signal computer card is specifically made to acquire and process the ground vibration signals using sensitive MEMS based accelerometers and high speed processor.
The software for MSS consists of two modules, namely:
a) Signal processing embedded system software running in processor
b) User interface software running in PC
Signal processing software running in processor is embedded system software, which acquires and processes the accelerometer data corresponding to ground vibrations and evokes the early warning alarm. User interface software running in PC provides the user interface facilities for communicating with MSS.
DSP software
The main objective of this program is to acquire 2 seconds real-time PWM data of vertical and horizontal channels of ADXL202 accelerometers and process the ground acceleration data to detect P-wave and S wave. The Richter magnitude (Ml) and epicenter distance of possible earthquake is calculated using early warning algorithm. If the estimated Richter scale magnitude crosses the pre set threshold (Ml) then a warning is generated. Five sub modules namely, Initialization routines, Set up parameters, Data Acquisition, Data Processing and Warning generation 125 are included which are designed around serial and timer interrupts of DSP. Data Processing is done in every 2 second. Serial interrupt is active when the user wants to send initialization parameters from local computer to MSS. Timer interrupt is active whenever the accelerometer PWM data is received at the timer ports of the processor. Data acquisition is initiated based on timer interrupt. Data processing is done in the main body of the program. Initialization routines take care of the initial MSS health checkups, Set up parameter sub module looks after the receipt of user selectable initialization parameters and passing of parameters to processor. Warning generation module evokes the early warning if Data processing module detects an event above the threshold. The flow chart of DSP software is given in figure 6.

Sub module 1-Initialization routines
Immediately after power up initialization module takes the control of code execution and checks the health status of MSS by inspecting the DSP interfaces with internal memory, external memory and external interfaces. The result of the initialization test is displayed by LED 2 to 3 of MSS using the flag I/O pins (FLAG 0-1). At the end of this module serial interrupt is enabled. The status of ConfigEdit flag is checked to either run with default values of set up parameters or accepting values from PC.
Sub module 2-Set up parameters
User can send the initialization parameters like values for Threshold strong motion, lower value of Tp, Upper value of Tp, S-wave detection factor, Range for processing, Valid count for frame validation and Threshold value of Ml to MSS from local computer user interface. After the execution of initialization routines sub module, processor enters to a for ever loop waiting for interrupts to come. Once the set up parameters are sent from local computer to the serial port of DSP in MSS, Interrupt Service Subroutine (ISR) for serial interrupt is evoked. In the ISR for serial data reception, the received ASCII characters are stored to a specific memory location after converting to float values and the same characters were sent back to local computer as an acknowledgement. If the character received at MSS is a valid one (valid means any of ASCII character 0 to 9, A to Z or decimal point) it is read from the receive buffer and stored in a receive array. This process is continued till all parameters are received from local computer. If the received character is 'Y' previous digits received are taken as the parameter for Threshold values for percentage strong motion. Similarly arrival of character 'X' indicates lower value of Tp, 'W indicates upper value of Tp,'V indicates S-wave detection factor, 'U' indicates range of processing after P-wave detection, 'T' indicates valid count for frame validation, 'K' indicates threshold for Ml. and arrival of character 'Z' for end of reception of parameters. The received parameters are then transferred to corresponding variables. Finally serial interrupt is disabled, timer interrupt is enabled and processor goes to forever wait loop waiting for timer interrupts
Calibration
If the character received from PC is 'C then MSS is delegated to calibration mode by setting calibration flag and ten second sensors data are acquired and from these values

Average static acceleration values in X(g) and Y(g) is calculated and these values are send to PC . The user can either accept these values so that DC offset removal is done by using these values or reject to run MSS with default DC offset removal values.
Sub module 3 - Data Acquisition
Data acquisition module consists of the ISR of timer interrupt. Horizontal and Vertical accelerometer PWM data are interfaced to Timer 1 and Timer 0 of DSP respectively. Whenever a PWM signal is received at the timer ports of DSP after enabling timer interrupt, the program will jump to the ISR of timer interrupts. In the timer ISR, the ON period (Tl) and total period (T2) of PWM signal taken from TCOUNT and TPERIOD registers respectively of timer. The duty ratio(Tl/T2) is calculated and stored in circular buffer, A TICKS (real-time system clock) counter is incremented on receipt of each set of PWM data. When the value of TICKS counter reaches twice the sampling frequency (equals to 2 second data), the program sequence goes to Data processing module.
Sub module 4 - Data processing
Data acquired by the Data Acquisition module are being processed in this module. The ground acceleration for both channels are calculated as follows: -
Acceleration in g = (Duty ratio - 0.5)/12.5 %;
Since 0 g corresponds to 50 % Duty Cycle of PWM signal of accelerometer The acceleration values are stored to acceleration buffer. Two second acceleration data samples of both x and y axis data is stored in process buffers and is read for further processing. The TICKS counter is reset to zero value. The following processes are carried out on the samples:-
a) DC offset removal
This process is necessary because in acceleration readings earth's g value and Og offset errors will act as a DC offset. DC shift is removed by subtracting either the average static g values calculated in calibration mode or the default values set by the program. This is done by using the following formula.
New value of ith sample value= ith g value -De value This is applicable for both x and y-axis.

b) Apply an FIR Low pass Filter with pass band end frequency 2Hz and stop band start frequency 5Hz to filter out extraneous noise.
c) Energy (Mean square value of one frame of data) is calculated before and after filtering. Energy passed through filter is calculated from this energy for both X and Y-axis.
d) Frame validation is done for X axis data for rejecting noise pick ups below noise floor and above the upper limit of accelerometer output saturation Validation is done for X-axis data alone since arrival of P-Waves to be detected from vertical accelerometer output.
e) Maximum g value and number of g that crosses the set value (Percentage strong motion) is found from both axis data.
f) Tp, Predominant Period is calculated from validated X-axis data samples, if energy passed is greater than preset energy threshold (default value 40%).
Algorithm for finding Tp
Where
TjP is the predominant period at time i, aj is the recorded ground acceleration , Ai is the smoothed ground acceleration squared, Jj is the jerk(time derivative of acceleration) squared and a is the smoothening constant .The value of a is 0.99.
g) The maximum value (Tp-max) of predominant period Tp, from 2-second samples
are found out. When Tp-max is between lower value of Tp and upper value of Tp, then
P- wave is detected following parameters are calculated.
mh, Body Magnitude = 7.0 * Logio(Tp-Max) + 5.9 Ml, estimated Richter magnitude = (1.5873*mh)-3.968253968 P-wave Confidence Factor is incremented whenever a P wave is detected.

Process to Detect S-wave
h) Once a P- wave is detected and the process will search for an S-wave from both X and Y-axis data. An S-wave is detected when current frame energy is >= twice that of detected P wave. When S- wave is detected, S-Flag is set and the time interval between P- wave and S-wave arrival is calculated. This time interval multiplied by 8KM will give the epicenter distance since there is typically a one second (approximately) separation between the P and S waves for every 8 KM traveled. This process is repeated for next frames of data.
1.4.Sub module S - Sending Results
Calculated parameters are sent from MSS to user interface running in local computer through the serial port of DSP. It includes the following: -
1) Result of Calibration
2) Maximum 'g' value of each frame.
3) Percentage Strong Motion
4) X (g) and Y (g) values
5) Maximum Tp.
6) Primary Confidence Factor
7) Secondary Confidence Factor
8) Ml, predicted Richter Magnitude
9) P-S time
User interface software
MSS User interface is a software developed using CVI(C for Virtual Instrumentation) to run in a local computer in order to communicate with MSS. The main functions carried out by this program are sending initialization parameters to MSS, displaying results of processing and to store the data and results of processing in files. Parameters like threshold strong motion, Tp-Lower (Lower value of Predominant Period), Tp-Upper (Upper value of Predominant Period), S-Wave Detection Factor, Range, Valid Count and Threshold Ml are send to the MSS via the serial port of PC/Laptop by pressing the SEND button in the user interface. Identification tag is attached with every parameter to distinguish between them .At the receiving side values of parameters are

stored in corresponding memory location for processing. MSS transmits the received values back to Laptop in order to make sure that the send parameters were reached in the destination properly. The CVI program running on the PC stores and displays the results sent by MSS.
When the CALIBRATE button is pressed the system enters in calibration mode by acquiring ten second acceleration data in both x and y-axis for calculating average static g values X (g) and Y (g) for DC offset removal. The calculated values are displayed in the acknowledgement window, user can either accept these values by pressing ACCEPT button or can run with default values set by MSS program by pressing RUN button.
MSS can be tested with simulated signals from programmable arbitrary wave form generator (Eg:- Agilent 33220A Model).In that mode DC offset removal is not required. LABTEST button is clicked for running with simulated signal. When RUN button is clicked serial interrupt is disabled, timer interrupt is enabled and processor goes to a 'for-ever-wait' loop waiting for interrupts and continuously acquires data and processes it and the results are transmitted through the serial interface. A real time data plotting facility is provided for displaying x and y channel acceleration data. Data and results are also stored in separate files, thus the user can analyze the results in offline. In the user interface 'P wave detected', 'S wave detected' message and calculated values of maximum 'g' in x and y channel data, percentage strong motion value, maximum Predominant Period (Tp), confidence factor for P-wave and S-wave, Body Magnitude of Earthquake (mh), and equivalent Richter scale magnitude(Ml) are displayed. Once P- wave is detected, MSS looks for consecutive detection of P-wave to set the confidence level and the arrival of S-wave. When S-wave is detected distance from Epicenter and confidence level is sent by MSS is displayed on Local computer. The flow chart of the program is as shown in figure 7.
User interface
The user interface part of this program includes the following components.

Numeric: - This control is used for displaying the values of maximum 'g' value, percentage strong motion, Ml, Ms, Tp-Max, Confidence factor for P-wave & S-wave and distance from epicenter.
Text Message: - Text Message is used to display strings of text. When a P-wave is detected it shows 'P-wave detected' message and when S wave is detected it shows 'S-wave detected' message.
Command Button: - This control is used to trigger an action.
SEND button associated with this is used for sending the parameters to MSS.
CALIBRATE button is used for displaying results after calibration.
ACCEPT button is used for accepting the results of calibration
RUN button is associated with running in continuous mode.
LABTEST button is used to run without DC offset removal
EXIT button is used for closing COM port, files and to quit the user interface
Ring Control: - Ring control is used to select from a group of items. Each item is stored in a ring control in the form of label/value pairs where the label is the text string displayed by the control when the item is selected and value is numeric or string value corresponding to that item. When this control is selected the user can set the parameter values for Threshold strong motion, Tp-Lower (Lower value of Predominant Period), Tp-Upper (Upper value of Predominant Period), S-Wave Detection Factor, Range, Valid Count and Threshold Ml.
Textbox Control: - This control is used for displaying acknowledgement and message to user.
Strip Chart: - Strip Chart control display graphical data in real time, consisting of one or more traces that is updated simultaneously. This control is used for displaying x and y axis acceleration data in real time. A typical user interface is as shown in figure 9.
MSS is a digital signal processing embedded system designed to acquire and process the ground vibration signals using sensitive MEMS based accelerometers and high speed DSP.

The hardware consists of a DSP based signal processing card (MSS Signal computer card) and a vertical accelerometer card (MSS Vertical card).
DSP will continuously acquire the signals from the accelerometers through its timers and will process it based on early warning algorithms for earthquake. A suitable electrical signal is generated when the processed magnitude crosses pre set threshold. The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. The processor used is ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS and 544Kbits internal SRAM.
Hardware-MSS Signal Computer card & MSS Vertical card
MSS Signal computer card specifically designed to acquire and process the ground vibration signals make use of sensitive MEMS based accelerometers and high speed DSP.
MSS Vertical card is based on sensitive MEMS based accelerometer to measure the vertical ground vibrations. The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. This card is fixed on the MSS signal computer card vertically so that the vertical vibrations of earth could be captured. The picture of MSS hardware is shown in figure 8. The block schematic of the MSS hardware is shown figure 1.
According to the block schematic of the MSS hardware is shown figure 1, the accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. It is a low power, low cost device having a resolution of 2mg at 60Hz.Two dual axis accelerometers are used to sense the horizontal and vertical vibrations. Horizontal accelerometer is fixed on MSS Signal computer card and vertical accelerometer is placed on MSS vertical card which is fixed vertically to MSS Signal computer card.
The processor selected is Analog Device's ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS. An external Synchronous DRAM memory with 256Mbits is provided for storing program data like look up tables etc. An EPROM with 256KB size is provided to keep the program code. JTAG interface is provided to communicate with DSP during the software development cycle. Reset

circuitry ensures the proper processor reset to a known state and ensures the execution of code from the hardware vectored address of program memory. The external communication interface of MSS is realized through the dedicated serial port of DSP.
Accelerometer Selection
The accelerometer used is Analog Device's MEMS based +/-2g Dual axis ADXL202JE with duty cycle output. It contains a polysilicon surface-micro machined sensor and signal conditioning circuitry to implement open loop acceleration measurement architecture. It is a low power, low cost device having a resolution of 2mg at 60Hz. The major features of ADXL202JE are:-
• MEMS based 2-axis acceleration sensor on a single chip
• 5 mm x 5 mm x 2 mm ultra small chip scale package
• Direct interface to DSP timer ports via Duty Cycle output
• Band Width adjustment with a single capacitorO
ADXL202JE can measure both dynamic acceleration (E.g. vibration) and static acceleration (eg. gravity). Two dual axis accelerometers are used to sense the acceleration in three mutually perpendicular axes. The duty cycle (ratio of pulse width to period) of PWM output of the sensor (figure 4) is proportional to acceleration. DSP timer ports can directly decode the duty cycle outputs.
Acceleration in g = ((T1/T2) - 0.5)/12.5 %
0 g = 50 % Duty Ratio (T1/T2)
T2(sec) = RSET(^ )/125MQ Where RSET is an external resistor to the ADXL202 Accelerometer.
Typical Duty Cycle output of ADXL 202 is as shown in figure 4.
The MSS senses the arrival of primary (P) waves, the first waves to reach the detector,
and so provides adequate time to react before the arrival of more destructive shear (S)
and surface (Love and Rayleigh) waves. The MSS will filter out extraneous noise
components.
Accelerometer interface to DSP
The Pulse Width Modulated output of ADXL202 is directly interfaced to the timer
ports of Analog Devices' DSP ADSP 21065L.One sensor is placed on the MSS Signal

computer card which contains the DSP while the other sensor is fixed on the smaller MSS vertical card which is soldered vertically to the MSS Signal computer card. From the two accelerometers four channels are available-namely, Horizontal X axis, Horizontal Y axis, Vertical X axis and Vertical Y axis.
Processor- ADSP 21065L
The processor selected is ADSP 21065L which is a high performance floating point SHARC processor with 198 MFLOPS from Analog Devices. It has four independent buses for Dual data, instructions, and I/O fetch on a single cycle. The key features of ADSP21065Lare:-
• 32-Bit Fixed-Point Arithmetic; 32-Bit and 40-Bit Floating Point Arithmetic
• User-Configurable 544 Kbits On-Chip SRAM Memory
• 66 MIPS, 198 MFLOPS Peak, 132 MFLOPS sustained performance
• Two External Port, DMA Channels and Eight Serial port
• SDRAM Controller for Glueless Interface to Low Cost external memory (@ 66 MHz)
• 12 Programmable I/O Pins and Two Timers with Event Capture Options
• Serial Ports with independent Transmit and Receive Functions
• 3.3 Volt Operation
Internal SRAM memory size is 544Kbits and an in built external memory interface is provided. Signal from each axis of accelerometer is acquired and processed by the DSP through its timer ports-Timer 0 and Timer 1. DSP will continuously acquire the signals from the accelerometers through its timers and will process it based on early warning algorithms for earthquakes.
DSP-External Memory interface
ADSP 21065L has in built SDRAM controller by which glue less interface of SDRAM (Synchronous Dynamic RAM) is achieved. An external Synchronous DRAM memory with capacity of 128Mbits from Micron Technology (Part No.MT48LC8M16A2TG-75) with 133MHz maximum refresh rate is used for storing program data like look up tables etc. and processed outputs of the sensor which crosses the threshold. Two SDRAMs are used so that a total memory capacity of 256Mbits is ensured.

An EPROM with 256KB size is provided to store the program code. This EPROM can be made as the boot device by hardware selection provided. Memory size selection jumpers are provided so that the EPROM size can be increased up to 1MB with same pin out if required.
Setting the BSEL input of DSP high and the BMS input low selects EPROM booting through the external port. The data pins of the byte-wide boot EPROM is connected to DATA7-0 of DSP. The lowest address pins of the processor are connected to the EPROM's address lines. The EPROM's chip select is tied to BMS and its output enable to RD signal of DSP
DSP Clock circuitry
Stabilized clock input to DSP and external devices are ensured using a programmable clock generator having integrated phase-locked loop and frequency select option. CY2292F from Cypress semiconductors is a 3.3V EPROM programmable clock generator with input crystal frequency varies from 10MHz-25MHz and programmable output clock frequency from 76.923 kHz to 66.6MHz. In this design an external 12MHz crystal is used as the input source. CY2292F is programmed to output 33MHz (DSP_CLK1) and 66MHz (DSP_CLK2) clock outputs
DSP Reset Circuitry
Reset circuitry for DSP ensures the proper processor reset to a known state and ensures the execution of code from the hardware vectored address of program memory. 3.3V Reset IC from Dallas Semiconductors, DS1233A, is used as the Reset IC which provides a 350ms Reset Pulse during power on reset as well as externally forced reset conditions.
DSP JTAG Interface
JTAG interface is provided to communicate with DSP during the software development cycle using Analog Devices' USB based In Circuit Emulator. Compiled output of codes can be downloaded to the internal memory of DSP through JTAG interface. JTAG based In Circuit Emulator helps to debug the software codes running in DSP.

External Communication Interface
The external communication interface of MSS is realized through the dedicated serial port of DSP. RS232 link is provided for communicating with local computer and an optional RF/optical fiber link can be added for communication with other MSS and central computer.
Using the synchronous serial ports (SPORTS) on the SHARC DSP, it is possible to implement a full-duplex, asynchronous serial port to communicate with UARTs, EIA-232 (RS-232) devices with minimal software overhead (less than 1 MIP for full-duplex, 115,200 bps operation)
The prototype of MSS was extensively tested with simulated signals at CDAC lab and in simulated vibration environment at ERTL lab, Trivandrum for validating the Early Warning algorithm. MSS was also subjected to noisy environment tests to prove that false alarms due to local noise sources are eliminated.
MSS was also tested using recorded strong motion seismograms available in public domain. We visited and contacted through mail to the following sites www.pnsn.org, www.anss.om, www.usgs.gov etc. We downloaded recorded digital data for strong motion earthquake from above mentioned sites and tested MSS. We also tested MSS in following environments:-
• Using the vibration test facility at ERTL, Thiruvananthapuram with frequencies 1Hz, 2Hz , 3Hz , 5Hz and 10Hz with acceleration as low as 20millig. The noise pickups at lower range of vibration table limited the test for low value of acceleration and lower frequency of vibration
• Using simulated PWM signal corresponding to P-S wave using Agilent's arbitrary wave form generator
• At quarry during rock blasts
• Near railway track when train passes
• Near National High-way during heavy vehicle traffic
By these tests the early warning algorithm is validated. Tests conducted near noise sources proved that MSS is not generating false alarms.

The MSS has to be installed properly for field applications in such a way that the natural frequency of the mechanical set up is well outside the band of P wave and S wave and the ground vibrations are coupled to MSS with minimum attenuation. Also the possible entry of ambient noises such as vehicle traffic, wind, rain etc. to MSS setup should be minimized.
For permanent fixture of MSS following setup can be used.
The structure may consist of an L-shaped platform fixed to a concrete vault built on hard soil or rock as shown in figure 5. MSS is fixed firmly on the mounting platform. To prevent noise from local sources like vehicle traffic guard the vault using loose sand .MSS should also be protected from wind, dust, sun and rain with proper roof and cover.
The current prototype of MSS is designed for two-axis measurement of seismic signals. Earthquake early warning system requires three-axis measurement of seismic signals. Therefore two MSS modules have to be used. The fixing arrangement of two MSS modules is as shown in figure 5. One MSS module (consists of MSS Signal Computer Card and MSS Vertical card) has to be fixed horizontally. The other MSS module (consists of MSS Signal Computer Card and MSS Vertical card) has to be fixed vertically.
MSS detects the pre-cursor of a strong earthquake and generates an early warning signal as follows.
a) Generate an Electrical Early warning signal based on the characteristics of P wave only.
b) Generates a warning based on the analysis of P-S complex waves.
MSS can also be used as a low cost low sensitivity Seismograph. MSS Performance Characteristics:
• Generate warning signal for Magnitude> 5.5 (can be set)
• Fast processing: less than 50 milliseconds (can be reduced)
• Operation: 3.3V, 0°C to 50°C (can be improved)
• Power consumption: 600milli watts (can be reduced)
• Time taken to generate Early warning: less than 2.5 sec from P-wave arrival (can be reduced)

The following examples are provided purely by way of exemplification and illustration and should not be construed to limit the scope of the invention in any manner.
The invention is further elaborated with the help of following examples. However these examples should not be construed to limit the scope of the invention.
Examples
Determination of Expected Early Warning Time
In the following two examples, the Applicants have made estimated time period which would be available for people at the target site to prepare for the earth quake after receipt of the early warning from the Micro-Electro-Mechanical Systems based electronic earthquake early warning device of the present invention.
Example 1: When MSS is placed away from Target station
When an earthquake occurs, energy radiates in all directions as three types of seismic waves called primary, secondary and surface waves. The energy of primary waves (P-waves) travels through the earth as a sequence of back and-forth vibrations parallel to the direction of propagation of the seismic waves. Secondary waves (S-waves), also called shear waves, are transverse in nature. . The principle of P and S wave separation is as follows.
Since the P waves travel approximately 1.7 times faster than the S waves, the greater the distance from the focus of an earthquake, the greater would be the time elapsed between the P and S waves. There is typically a one-second separation between the P and S waves for every 8 km traveled. Surface waves (Rayleigh waves and Love waves) are the slowest and most destructive among the three.
Assume MSS is placed fD' Kms away from the target station to be protected. Let 'd' be the distance of MSS from the epicenter as shown in figure 2.
Time taken to generate warning signal at MSS from the arrival of P-wave = 2.2 sec Time taken to transmit warning to target station = 0.3 sec.
Time taken to receive early warning signal at target station from the arrival of P-wave at MSS = P = 2.2 + 0.3 = 2.5 seconds.

The velocity of the P-Wave depends on the type of soil .It varies from 0.3km/s in
aerated soil to 1 lkm/s at the centre of the earth. In the upper mantle the velocity of the
P Wave is 8km/s. Following derivation assumes that average velocity of P-Wave is
8km/s.
Early Warning Time [TE] = (0.21D + 0.085d - P) seconds
Hence the Early Warning Time [TE] = (0.21D + 0.085d - P) seconds.
If MSS is installed 20KM from station to be protected then D=20KM, then for various 'd' Early warning time can be calculated as follows and the same is tabulated in Table 1:

Example 2: When the MSS is located at the Target Site
Assume MSS is placed at the target station to be protected. Let 'd' be the distance of MSS from the epicenter as shown in figure 3.
Time taken to generate warning signal at MSS from the arrival of P-wave = P= 2.2 sec. Hence the Early Warning Time [T E] = (0.0850d-P) seconds.

For various'd' Early warning time can be calculated as follows and the same is tabulated in Table 2:
TABLE 2: EARLY WARNING TIME AVAILABLE

Industrial Application:
• In Control Systems for industries handling flow of hazardous and flammable material through pipes.
• Seismic Fence for Atomic and Thermal Power stations.
• As low sensitivity low cost Seismograph.
• It provides opportunities to automatically trigger measures such as shutdown computers; reroute electrical power; shutdown disk drives; shutdown high precision facilities; shutdown airport operations; shutdown manufacturing facilities; stop trains; shutdown high energy facilities; shutdown gas distribution; alert hospital operating rooms; open fire station doors; start emergency generators; stop elevators in a safe position; shutoff oil pipelines; issue audio alarms; shutdown refineries; shutdown nuclear power plans; shutoff water pipelines; maintain safe-state in nuclear facilities.

We Claim:
1. A method to generate earthquake early warning signal using MEMS based
seismic sensors (MSS), wherein said method comprising steps of;
a. positioning one MSS each horizontally and vertically onto a L-shaped
platform at a predetermined distance from target station,
b. detecting and processing P & S wave signals by said MSS, and
c. sending early warning signal to the target station.
2. The method as claimed in claim 1, wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock.
3. The method as claimed in claim 1, wherein detecting P & S waves using accelerometer.
4. The method as claimed in claim 3, wherein the accelerometer generates Pulse Width Modulated (PWM) output.
5. A system to generate earthquake early warning signal using MEMS based seismic sensors (MSS), wherein said system comprises;
a. MSS positioned both horizontally and vertically onto a L-shaped
platform to process P & S wave signals at a predetermined distance from
target station, and
b. means to send early warning signal to the target station.
6. The system as claimed in claim 5, wherein the L-shaped platform is fixed onto a concrete vault built over hard soil or rock.
7. The system as claimed in claim 5, wherein the sensor is an accelerometer.
8. The system as claimed in claim 5, wherein the MSS comprises a Digital Signal Processor (DSP).
9. The system as claimed in claim 5, wherein the system comprises to store the processed data.

Documents:

1061-CHE-2005 AMENDED PAGES OF SPECIFICATION 27-02-2012.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 10-03-2011.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 17-09-2010.pdf

1061-CHE-2005 POWER OF ATTORNEY 27-02-2012.pdf

1061-CHE-2005 AMENDED CLAIMS 27-02-2012.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 02-03-2012.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 03-11-2010.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 06-05-2011.pdf

1061-CHE-2005 CORRESPONDENCE OTHERS 27-02-2012.pdf

1061-CHE-2005 CORRESPONDENCE. OTHERS 27-02-2012.pdf

1061-CHE-2005 CORRESPONDENCE-OTHERS 09-09-2009.pdf

1061-CHE-2005 OTHER PATENT DOCUMENT 22-10-2009.pdf

1061-che-2005-abstract.pdf

1061-che-2005-claims.pdf

1061-che-2005-correspondnece-others.pdf

1061-che-2005-correspondnece-po.pdf

1061-che-2005-description(complete).pdf

1061-che-2005-description(provisional).pdf

1061-che-2005-drawings.pdf

1061-che-2005-form 1.pdf

1061-che-2005-form 3.pdf

1061-che-2005-form 5.pdf

1061-che-2005-form 9.pdf

« Previous Patent

Next Patent »

Patent Number

251809

Indian Patent Application Number

1061/CHE/2005

PG Journal Number

15/2012

Publication Date

13-Apr-2012

Grant Date

04-Apr-2012

Date of Filing

04-Aug-2005

Name of Patentee

CENTRE FOR DEVELOPMENT OF ADVANCED COMPUTING (CDAC)

Applicant Address

(AN AUTONOMOUS SOCIETY OF MINISTRY OF INFORMATION TECHNOLOGY, GOVERNMENT OF INDIA) THIRUVANANTHAPURAM UNIT,P.B. NO 6520, VELLAYAMBALAM, THIRUVANANTHAPURAM 695 033, KERALA, INDIA

Inventors:

#	Inventor's Name	Inventor's Address
1	G. REGHUNATHAN NAIR	CDAC,THIRUVANANTHAPURAM UNIT,P.B. NO 6520, VELLAYAMBALAM, THIRUVANANTHAPURAM 695 033, KERALA, INDIA
2	K.R. RAJESH	CDAC,THIRUVANANTHAPURAM UNIT,P.B. NO 6520, VELLAYAMBALAM, THIRUVANANTHAPURAM 695 033, KERALA, INDIA

PCT International Classification Number

G01V 1/00

PCT International Application Number

N/A

PCT International Filing date

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1			NA